A new Perspective on Dichotomies in Socionics - Pyramid Diagrams, Draft

**Lao Tzunami** · 12-09-2017, 06:58 AM

I had a Eureka moment and realized I could do the same analysis I've been doing with the small groups on the dyad and type level. I didn't want to have to figure out all of those equations, so I wrote a program to do it for me \(^u^)/

This is written in Python, which you can run in browser [link]. Otherwise, here is the code

The program as posted is set to output excel cells for my dichotomy calculator. Changing a few things around [link], this is the output in Reinin notation:

E

N

T

P

EN

ET

EP

NT

NP

TP

ENT

ENP

ETP

NTP

ENTP

Now I just have to figure out how to factor in information metabolism and the higher order objects, like small groups, dyads and types. I might also want to put things in both positive and negative terms, like someone who is an introvert necessarily is not and extrovert.

**Lao Tzunami** · 01-07-2018, 07:47 PM

Right
Click
..+
Loop

.
.
.
.

.
.
.

.
.
.
.
.
.
.
.
.
.
.

**Elstner** · 03-24-2018, 10:22 PM

Thanks so much for creating these diagrams and all the information. I am using this extensively! Can you please tell what is the name program you used for creating the pyramid which are on youtube videos? Thanks in advance!

**Lao Tzunami** · 03-24-2018, 11:54 PM

Originally Posted by Elstner

Thanks so much for creating these diagrams and all the information. I am using this extensively! Can you please tell what is the name program you used for creating the pyramid which are on youtube videos? Thanks in advance!

Thanks! I put in all the work just for people like you <3

I used 3DS MAX for the modeling and rendering (series of individual .png images), and VEGAS 14 for compiling the image sequences into 60fps HD video. If you have a student email, you have access to all autodesk products, but if you've never used a 3d modeling program, 3DS MAX has a steep learning curve. Adobe After Effects is the industry standard program for post processing, I'm just using VEGAS 14 because I got it for $20 in a humble bundle.

...Unless you're also talking about the graphics in the article and excel program lol, I made those with the shape tools in word and excel ;.D

**Lao Tzunami** · 04-13-2018, 03:54 AM

I ran 20,000 random trials through my dichotomy analyzer algorithm and got this very narrow distribution curve. The harmony score, which measures how well the inputed traits theoretically agree with each other, can range from -35 to +91. In most science, 5% error is the gold standard for research, which translates to any score between +2.2 to +91. If I rate myself with the reinin dichotomies, I get a score of +35, which is well outside the random range.

I look forward to eventually putting together a test and applying this algorithm to see how the actual scores differ from purely random scores. If it is well outside the random range, it is empirical proof that the test, and by extension socionics, works.

Full Range (bin size .5):
Harmony Score Histogram Full Range.jpg

Zoomed on peak (bin size is .1):
Average Score distribution.jpg

**nyessss** · 05-07-2018, 11:41 PM

It will be significant

**ouronis** · 05-09-2018, 01:00 AM

How's your harmony score calculated? I find it strange that the random data clumped up so tightly around 0.

**Lao Tzunami** · 05-09-2018, 08:32 PM

Originally Posted by ouronis

How's your harmony score calculated? I find it strange that the random data clumped up so tightly around 0.

This curve was generated by randomly selecting all 15 dichotomy traits on a 5 likert scale (an even chance of –1, –.5, 0, +.5, +1), and recording the results 20,000 times.

Since then, I've run 100,000 cycles with an 11 likert model, and included the output in the new version of the excel file [link]. If you open it, remember to download, open in excel and enable editing and content. On the "Input" tab are all the options, and you can click the random button to see how it works. I've added comments that will tell you about the buttons if you hover your mouse over them. You can also run your own experiment on the 6th "Random Experiment" tab, just start low to see how your computer handles then number crunching.

All the equations are on the "(Harmony Calc)" tab, if you want to figure out how it works.

This is the result of how I rate myself. You can see even though there is variation, there are clear column stripes. If you put in a random result, the field will look like tv static.

**nyessss** · 05-16-2018, 07:56 PM

11

**Lao Tzunami** · 09-15-2018, 03:45 AM

Ever since @thehotelambush showed be there are additional dichotomies (we've been calling Tencer dichotomies), I've been trying to wrap my head around them. I just had a break through when I built a program to combine the Reinin and Tencer dichotomies, and it turns out they only generate 8 additional dichotomies! I'm in shock, that is so much more manageable than what I was expecting

https://repl.it/@ajsindri/Group-Generator

The new dichotomy space is the 8 yellow tables. All spaces share the black dichotomies in common.

**Bertrand** · 09-15-2018, 03:59 AM

looks like tetris

**Lao Tzunami** · 09-15-2018, 08:47 PM

After an existential crisis, I've decide to redo all my 4x4 tables to show Jung's type compass cycle. These are the changes:

In this new convention, here are all 32 structural type dichotomies.

I'll start updating all of my programs, article, charts, videos, ect, but I'm in the middle of a college semester, so it might be a while. The old tables still work great, but the new ones should be easier to see patterns with.

**Bertrand** · 09-16-2018, 06:06 AM

did you know strat has her own extension of reinin dichotomies? its interesting she has words for all the new combinations and works them into her narrative too. I like "smoothies" which is one of them

**Lao Tzunami** · 09-16-2018, 03:21 PM

Originally Posted by Bertrand

did you know strat has her own extension of reinin dichotomies? its interesting she has words for all the new combinations and works them into her narrative too. I like "smoothies" which is one of them

Naw, I don't really care about the definitions. I'm trying to work out the math so everyone can come to the table and figure out what details work and which don't. There are so many different ideas and theories, that level of synthesis is beyond the ability of a person, but we can do it as a community with the right structure.

**Lao Tzunami** · 09-26-2018, 11:28 PM

This is some more exploratory work trying to reconcile the Reinin and Tencer interpretations and trying to see if a Dihedral Four Group can be generated from the compass dichotomies. I know these are complicated, but the whole point is try to integrate as much information together as possible to see if there is an underlying shape that makes sense of the whole pattern. Instructions are in the 3rd image. Feel free to ask questions if you feel lost.

**Lao Tzunami** · 10-13-2018, 09:22 PM

This is a matrix for every combination of orbital, Reinin, Tencer and cross dichotomies. The color cells are + and the white cells are - . You can view with google, but if you download and open in Excel, the first row and column are always displayed, which is almost an essential feature if you want to reference this table. Also, the top right of each matrix is commented with the dichotomy traits.

https://drive.google.com/open?id=1vYppKMzPHpR7MTJGzrnGi4AP1yTnr-w3

**Lao Tzunami** · 02-12-2019, 04:23 AM

This is how the harmony value could be used to test socionic theory.

Thread: A new Perspective on Dichotomies in Socionics - Pyramid Diagrams, Draft

Thread Tools

Display

Error correction on all structural levels

Program name

Tags for this Thread

Posting Permissions